As someone with 17 years of experience in working with real world data, I can say that we’re in one of the most exciting times for generation of clinico-genomic datasets and an ability to advance precision medicine. Back in 2008, cancer centers were just switching from paper patient records to electronic health records, and we were suddenly able to aggregate EMR data to learn from a vast number of patients undergoing standard of care vs. characterizing outcomes from standard of care in small cohorts from chart review data. The potential strength of real world EMR data at scale characterizing heterogeneous treatment approaches with outcomes was only in its infancy. Previously, RWD insights were primarily gained from analysis of claims data that lacked important clinical detail. Through my collaborations with early innovators in the real world data sphere, we realized we could evaluate treatment sequencing, multiple grouping variables, and other predictors to gain insights that couldn’t be included in randomized controlled clinical trials but could be gathered from patients in the real world.

Dr. Stepanski and his family

The importance of molecular genomic data in oncology has exploded over the past 20 years. Obtaining broad molecular genomic data early in the treatment of advanced cancer in order to deliver optimal care became obvious to me from experience in supporting the pivotal crizotinib trial in patients with non-small cell lung cancer with ALK-positive tumors . At that time, it was believed that the base rate of the required ALK mutation was about 5-7% (now understood to be even lower). To find eligible study patients, we had to consent 20 patients and perform molecular testing on tumor tissue in order to find one eligible patient. This meant having tough conversations with 19 out of 20 patients who weren’t eligible and were waiting for treatment while their tumor grew. The same hurdle existed post-approval in establishing treatment eligibility for patients in clinical care. However, when the ALK result was known during first-line treatment, having a second-line treatment available that targeted a patient’s rare mutation entirely changed the dynamic in presenting this opportunity to the patient. The healthcare team could present a treatment known to be effective in treating a patient with a rare mutation at the time of disease progression without waiting – welcome news for the patient.

But even today, there are patients with NSCLC who are not tested for ALK or other new biomarkers that could improve their outcomes with targeted therapy aimed at their specific mutational profile. Since that trial there have been many more examples of new biomarkers and targeted agents that significantly improve treatment outcomes in oncology indications. We still have much work to do in the area of biomarker discovery that will drive insights, including an enormous opportunity for biomarker discovery in non-oncology therapeutic areas.

Today, genomic data gathered from real world patients is at a similar junction as RWD was back then. In Oncology, much of the genomic data we have comes from NGS or other omics testing which is completed as part of clinical care workflows. These data, while having some limitations, have resulted in a proliferation of targeted therapies over the last decade which have partially been responsible for lowering death rates for those with cancer. Access to research grade molecular genomics data including WGS deep omics are significantly limited in areas outside of oncology, like autoimmune and cardiometabolic disease. This is primarily because the data we have are limited to what we can capture as part of standard of care and the rate of genomic testing as standard of care is enormously reduced in non-oncology indications.

Meanwhile, the technology of advanced molecular genomics is moving so quickly that we can’t rely on last year’s omics data for drug research and development. We need to create robust genomics data sets with linked phenotypic data that combine the strengths of newer sequencing technology with deep patient clinical history in order to get to the vision of precision medicine that is possible with research grade genomics data sets.

That’s why at Ovation we’ve made a commitment to provide the richest possible genomic data for discovery including WGS and WES for use in drug discovery and development to push beyond the current limits of standard of care genomic data in RWD datasets. By having consented specimens, we are able to create multi-omics datasets based on the latest technology rather than rely on clinical testing done in the past that often does not include the depth and breadth of -omics as needed in the current time to drive discovery. With ‘-omics on demand’ linked with fit for purpose phenotypic data, we have the potential to identify and validate many more biomarkers than are currently available to inform treatment decisions for patients.

This is what I love about Ovation’s data – the ability to create genomic data from our specimens and then link to the rapidly growing number of phenotypic datasets to optimize the analytic value of the clinico-genomic dataset. Ovation is agnostic to tokenization mechanisms and source of phenotypic data linkage. This allows us to link the best phenotypic data to our genomic data and ultimately, impact patient health and outcomes more rapidly. I’m thrilled to work with the top scientific and technical teams at Ovation to achieve this mission.

Dr. Stepanski recently oversaw operations of Real World Data Products at ConcertAI/Vector Oncology and previously managed an oncology clinical trials network for over ten years, as well as a full-service CRO managing oncology clinical trials.

Get in Touch

If you want to learn more about our mission to advance the development of precision medicine by enabling the use of molecular and human genomic data at scale, get in touch with our team.